The long-term effectiveness of universal, selective and combined prevention for alcohol use during adolescence: 36-month outcomes from a cluster randomized controlled trial.

Abstract

To compare the long-term universal outcomes of the Climate Schools programme, the selective preventure programme and their combined implementation to standard substance use education in reducing the uptake of alcohol use, engagement in binge drinking and alcohol-related harms over a 3-year period. A cluster-randomized controlled trial. Substance use prevention programmes delivered in Australian secondary schools. Students from 26 Australian secondary schools (n=2190), mean age at baseline 13.3 years (standard deviation=0.48), 57.4% male. Schools were recruited between September 2011 and February 2012. Schools were block-randomized to one of four groups: universal prevention (climate; 12×40-minute lessons); selective prevention (preventure; 2×90-minute sessions); combined prevention (climate and preventure; CAP); or health education as usual (control). The climate intervention delivered 12×40-minute lessons aimed at reducing alcohol and cannabis use and related harms. The preventure intervention delivered 2×90-minute group sessions to high-risk students. The CAP group implemented the climate programme to the entire year group and the preventure programme to the high-risk students. Participants were all consenting 8th grade students (in 2012) assessed at baseline, post-intervention (6-9months post-baseline) and at 12, 24 and 36months post-baseline on measures of alcohol use, knowledge and related harms. Primary outcomes were alcohol use, binge drinking (five or more standard drinks) and alcohol-related harms, obtained from all students regardless of whether or not they received intervention. Intervention effects at 36months post-baseline were estimated from generalized multi-level mixed models using data from all time-points and accounting for school-level clustering. Exploratory analyses examined intervention effects among low- and high-risk adolescents. Compared with students in the control condition, students in the climate, preventure and CAP groups demonstrated significantly slower increases in their likelihood to drink any alcohol [odds ratio (OR)=0.64, 95% confidence interval (CI)=0.50-0.82 for climate; OR=0.55, 95% CI=0.43-0.71 for preventure and OR=0.67, 95% CI=0.53-0.84 for CAP] to engage in binge drinking (OR=0.60, 95% CI=0.44-0.82 for climate; OR=0.59, 95% CI=0.44-0.80 for preventure and OR=0.68, 95% CI=0.51-0.92 for CAP) and to experience alcohol harms (OR=0.63, 95% CI=0.49-0.82 for climate; OR=0.55, 95% CI=0.43-0.71 for preventure and OR=0.64, 95% CI=0.50-0.81 for CAP). There was no strong evidence that the combined approach showed advantages over universal prevention. The direction and magnitude of effects were consistent in low- and high-risk adolescents. The universal Climate Schools programme and the selective preventure programme were effective in reducing alcohol consumption and alcohol problems compared with standard Australian health education, when trialled individually and together over a 3-year period.